opioid substitution therapy in manipur and nagaland, north-east india: operational research in...

RESEARCH Open Access

Opioid substitution therapy in manipur andnagaland, north-east india: operationalresearch in actionGregory Armstrong1*, Michelle Kermode1, Charan Sharma2, Biangtung Langkham3, Nick Crofts1

Abstract

Background: There is good evidence for the effectiveness of opioid substitution therapy (OST) for injecting drugusers (IDUs) in middle and high-income countries but little evidence regarding the provision of OST by non-government organisations (NGOs) in resource-poor settings. This paper reports on outcomes of an NGO-based OSTprogram providing sub-lingual buprenorphine to opiate dependent IDUs in two north-east Indian states (Manipurand Nagaland), a region where conflict, under-development and injecting of heroin and Spasmoproxyvon (SP) areongoing problems. The objectives of the study were: 1) to calculate OST treatment retention, 2) to assess theimpact on HIV risk behaviours and quality of life, and 3) to identify client characteristics associated with cessationof treatment due to relapse.

Methods: This study involves analysis of data that were routinely and prospectively collected from all clientsenrolled in an OST program in Manipur and Nagaland between May 2006 and December 2007 (n = 2569, 1853 inManipur and 716 in Nagaland) using standardised questionnaires, and is best classified as operational research. Thedata were recorded at intake into the program, after three months, and at cessation. Outcome measures includedHIV risk behaviours and quality of life indicators. Predictors of relapse were modelled using binary logisticregression.

Results: Of all clients enrolled in OST during the month of May 2006 (n = 713), 72.8% remained on treatment afterthree months, and 63.3% after six months. Statistically significant (p = 0.05) improvements were observed inrelation to needle sharing, unsafe sex, incidents of detention, and a range of quality of life measures. Greaterspending on drugs at intake (OR 1.20), frequently missing doses (OR 8.82), and having heroin rather than SP as themost problematic drug (OR 1.95) were factors that increased the likelihood of relapse, and longer duration intreatment (OR 0.76) and regular family involvement in treatment (OR 0.20) reduced the likelihood of relapse.

Conclusion: The findings from this operational research indicate that the provision of OST by NGOs in the severelyconstrained context of Manipur and Nagaland achieved outcomes that are internationally comparable, andhighlights strategies for strengthening similar programs in this and other resource-poor settings.

BackgroundOpioid substitution therapy (OST) is an evidence-basedintervention for opiate dependant persons that replacesillicit drug use with medically prescribed, orally adminis-tered opiates such as buprenorphine and methadone.OST reduces HIV risk behaviours and harms associatedwith injecting (such as abscesses, septicaemia and

endocarditis), overdose and participation in criminalactivity, thereby improving the quality of life and healthof injecting drug users (IDUs) [1-6]. It is endorsed byUNAIDS, UNODC and WHO as part of a comprehen-sive package of nine core interventions for IDUprograms that collectively maximise impact for HIVprevention and treatment [7]. However, most of theevidence for OST effectiveness has been generated inmiddle and high-income countries where programs aremostly located in dedicated healthcare settings; evidenceregarding the outcomes of OST programs in low-

* Correspondence: [email protected] Institute for Global Health, University of Melbourne, Victoria, AustraliaFull list of author information is available at the end of the article

Armstrong et al. Harm Reduction Journal 2010, 7:29http://www.harmreductionjournal.com/content/7/1/29

© 2010 Armstrong et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.

mailto:[email protected]

income countries where OST is often provided in grass-roots settings such as drop-in-centres, is limited [3,8].There are an estimated 106,000-223,000 IDUs in India,of whom only 5% are currently receiving OST, which ismostly delivered by community-based services [9,10].There is a real need for evidence regarding outcomes ofOST provision in India in order to strengthen the casefor scaling up of services.This paper reports on outcomes of an OST program

providing buprenorphine to opiate dependent IDUs,delivered by non-government organisations (NGOs) inthe north-east Indian states of Manipur and Nagaland.These states make up a region geographically isolatedfrom the rest of India, and characterised by multiplesources of conflict including a longstanding civil insur-gent struggle, poverty and unemployment. Approxi-mately 2% of the population in Manipur and Nagalandinject drugs, [11] most commonly heroin and Spasmo-proxyvon (SP, a synthetic opioid analgesic that containsdextropropoxyphene, dicyclomine hydrochloride andparacetamol). As a consequence, Manipur and Nagalandare the two states with the highest HIV prevalence inthe country [11]. Both the epidemic and the response toit are more mature in Manipur, where sentinel surveil-lance data indicates that during the late 1990s HIV pre-valence among IDUs approached 80% [12]. By 2007,HIV prevalence among IDUs was much reduced being18% in Manipur and 1.9% in Nagaland [13]. Theresponse to HIV and injecting drug use in this geo-poli-tically complex environment was punitive and coercive,but harm reduction interventions such as needle andsyringe exchange programs and condom distributionhave been government policy since the mid 1990s [14].Project ORCHID (Organised Response for Comprehen-

sive HIV Interventions in the Districts of Nagaland andManipur) is a Bill & Melinda Gates Foundation-fundedHIV prevention project that has been working in selecteddistricts of Manipur and Nagaland since 2004. It supportslocal partner NGOs to deliver a range of harm reductioninterventions in rural and urban settings. In 2006, ProjectORCHID initiated a buprenorphine-based OST programdelivered by 11 local partner NGOs, initially with fundingfrom the United Kingdom government’s Department forInternational Developing (DFID), and subsequently fromthe National AIDS Control Organisation (NACO) andEmmanuel Hospital Association (EHA). The OST pro-gram is based in the community, operated out of drop-incentres. Sub-lingual buprenorphine is provided for regis-tered IDUs seven days per week, and is administered bytrained health care workers (mostly nurses) under thesupervision of medical doctors, following a standardisedprotocol. The program was initially rapidly over-sub-scribed and waiting lists were created. The program ismore fully described elsewhere [15].

During the DFID-funded period of the program (May2006 - December 2007) more detailed informationregarding characteristics of the clients and outcomes ofthe program were systematically collected as part ofroutine program monitoring. Analysis of these data wereundertaken in order to address the following objectives:1) to calculate OST treatment retention at 3, 6, 9 and12 months, 2) to assess the impact of OST on HIV riskbehaviours and quality of life, and 3) to identify clientcharacteristics associated with reason for cessation ofOST treatment.

MethodsStudy designThis study involves analysis of data collected routinelyduring the implementation of an OST program, and isbest classified as operational research, which can bedefined as “The search for knowledge on interventions,strategies, or tools that can enhance the quality, effec-tiveness or coverage of programmes in which theresearch is being done” (p.711) [16]. There is a strongconnection between program monitoring and evaluationand operational research. Study designs such as rando-mised controlled trials generate new knowledge aboutthe efficacy of interventions in a controlled environmentwith strict inclusion and exclusion criteria, whereasoperational research assesses effectiveness in routine set-tings that are far less controlled. The findings fromoperational research have direct and practical implica-tions for health care delivery [16].

Data collectionData were prospectively collected from all clientsenrolled in the OST program in Manipur and Nagalandbetween May 2006 and December 2007 (n = 2569, 1853in Manipur and 716 in Nagaland) at intake, threemonths after entry into the program, and at cessation oftreatment (regardless of the reason) using standardisedquestionnaires developed by the program. The question-naires were interviewer-administered by the NGO nurseor outreach worker, and took approximately thirty min-utes to complete. It was not always possible to conducta face-to-face interview with clients at cessation of treat-ment, especially if cessation was due to relapse, sowhere necessary and possible, relevant information wasdrawn from the client file.

Outcome measuresThe intake and three month follow-up questionnairescaptured self-reported information on socio-demo-graphic characteristics, drug use, HIV risk behaviours,and quality of life. At cessation of treatment additionalinformation was recorded regarding reason for cessation,family involvement during treatment, and adherence to


of 7

treatment. Reasons for ceasing OST were categorised as:completed the program (meaning that the clients hadwithdrawn from buprenorphine and had not returned totheir past pattern of drug use at the time of discharge);relapsed or involuntarily discharged (hereafter referredto as relapsed); and unknown reason for cessation.

AnalysisData were entered by the Project ORCHID monitoringand evaluation team using EpiInfo, and analysed usingSPSS version 18. The statistical tests used were Chi-square, t-test, and McNemar’s test, and statistical signifi-cance was calculated using two-tailed tests at the 95%confidence level. Clients who had ceased OST with anunknown reason (n = 281) were excluded from the ana-lysis, except when calculating OST treatment retentionand describing the client characteristics. In order tocalculate OST treatment retention at 3, 6, 9 and12 months, all clients commencing OST during May 2006(n = 713) were tracked over the subsequent 12 months.The impact of OST on HIV risk behaviours and qual-

ity of life was assessed by comparing changes betweenbaseline and three month follow-up measures. Resultswere differentiated by the programmatic status of clientsat the end of the data collection period i.e. completedthe program, relapsed, or still on OST.To determine factors associated with reason for cessa-

tion we identified all clients who had ceased treatmentwith a known reason for cessation (n = 895) i.e. thosewho had either completed the program or had relapsed.A binary logistic regression model was used to predictthe likelihood of relapse at cessation of treatment ratherthan completion of the program. Unadjusted odds ratioswith p-values less than 0.1 were considered eligible forthe multivariate model, and gender and age were alsoincluded. The forced entry procedure was used to entervariables in the model.

ResultsClient characteristicsTable 1 presents socio-demographic data for all clients atentry to OST disaggregated by state. In both Manipur andNagaland, clients were predominantly male and the major-ity had at least a high school level of education. Almosthalf reported being unemployed and the most commonsource of referral to OST was friends/peers. A small pro-portion of the OST clients in Nagaland (13.2%) werefemale sex workers. Ages ranged from 16 to 61 years inManipur with a mean age of 30.9 years. In Nagaland agesranged from 18 to 55, with a mean age of 30.0 years.There was variation in drug use between Manipur and

Nagaland; at intake most clients in Manipur reportedcommonly using heroin (90.7%) whilst in Nagalandapproximately equal proportions reported using heroin

and SP (63.1% and 68.3% respectively). Clients fromNagaland more commonly reported use of other drugsincluding alcohol (50.9%), Relipen (20.4%; combinationdrug containing similar ingredients to SP) and Nitrosun(26.4%; nitrazepam). The majority of OST clients inManipur identified heroin as their most problematicdrug (87.6%), while in Nagaland the most problematicdrug was evenly split between heroin and SP (50.2% and47.7% respectively).

OST treatment retentionOf all clients enrolled in OST during the month of May2006 (n = 713), 72.8% remained on treatment afterthree months, and 63.3% after six months (Table 2). Atthe end of one year, 50.8% were still on OST. Approxi-mately two-thirds (63.6%) had what can be defined as apositive outcome after one year i.e. 12.8% had com-pleted the program and 50.8% were retained on treat-ment. Slightly more than one-quarter (27.5%) hadceased treatment at the end of one year due to relapse,and the remaining 9% had ceased treatment with anunknown outcome.

Impact of OST on HIV risk behaviours and quality of lifeSubstantial improvements in self-reported HIV riskbehaviours were observed among clients retained onOST between intake and 3 months (Table 3). Therewere significant reductions in needle sharing and unsafesex. At intake one-quarter of clients reported sharingneedles in the past month compared to 2% or less afterthree months on OST. There was a significant decreasein the proportion of clients being jailed/detained. Reduc-tions in HIV risk behaviours were observed amongst allclients on treatment, even those clients who went on tocease OST due to relapse.There was a consistent and marked improvement

observed in the quality of life measures when intake iscompared with three months after enrolment (Table 4).Of the clients successfully followed-up at 3 months, theproportion reporting a good quality of life had risen byapproximately 40-50%. Other statistically significantimprovements in quality of life were also evident includ-ing increased attendance at social events, reducedfrequency of family conflict, and a reduction in work-related absenteeism amongst those with a job. Theimprovements in quality of life were observed amongstall clients on treatment, even those clients who went onto cease OST due to relapse. Notably, no statisticallysignificant changes were observed with respect to theproportion of clients who were employed.

Reasons for cessation of OST treatmentOf the 895 clients who ceased OST treatment duringthe data collection period, 57% (n = 510) left OST


of 7

because they had relapsed, and 43% (n = 385) leftbecause they had completed the program without areturn to their previous pattern of drug use at the timeof discharge.Binary logistic regression modeling was performed to

assess the relative impact of a range of factors on thereason for cessation (Table 5). The dependent variablewas reason for cessation i.e. relapse versus completionof the program. The model contained gender and age aswell as duration in treatment, most problematic drug,amount of money spent daily on drugs at intake, fre-quently missing more than two doses a week, and regu-lar family involvement in treatment. This modelexplained between 43.9% (Cox and Snell R square) and58.6% (Nagelkerke R square) of the variance in reasonfor cessation.Gender and age were not statistically significant pre-

dictors of reason for cessation. Five variables made astatistically significant contribution to the model; dura-tion in treatment, most problematic drug, money spentdaily on drugs at intake, frequently missing doses, and

regular family involvement in treatment. Greater spend-ing on drugs at intake, frequently missing doses, andhaving heroin rather than SP as the most problematicdrug were factors that increased the likelihood of cessa-tion due to relapse, and longer duration in treatmentand regular family involvement in treatment reducedthe likelihood of cessation due to relapse.Among the clients who ceased treatment, those who

reported heroin as their most problematic drug werealmost twice as likely to relapse compared to thosereporting SP. Clients who frequently missed more thantwo doses a week were almost nine times more likely tocease treatment due to relapse. Every additional monthspent in treatment reduced the risk of cessation due torelapse by 24%. Clients whose families were not regu-larly involved in their OST treatment were five timesmore likely to cease treatment due to relapse.

DiscussionThis study aims to contribute to the evidence-base forthe provision of OST by NGOs in northeast India, a

Table 1 OST client socio-demographic characteristics at intake (n = 2569)*

Demographic characteristic Manipur n (%) Nagaland n (%) Demographic characteristic Manipur n (%) Nagaland n (%)

Sex Education

Male 1775 (96.3) 598 (85.1) No education 103 (5.6) 49 (6.9)

Female 69 (3.7) 105 (14.9) Primary school 450 (24.3) 147 (20.6)

High school 444 (24.0) 243 (34.1)

Marital Status Undergraduate 548 (29.6) 146 (20.5)

Married 873 (47.1) 361 (50.5) Graduate and above 306 (16.5) 128 (18.0)

Single 876 (47.3) 320 (44.8)

Separated/divorced 67 (3.6) 23 (3.2) Occupation

Widowed 36 (1.9) 11 (1.5) Unemployed 874 (48.6) 323 (45.3)

Small business 335 (18.6) 72 (10.1)

Source of referral Government 118 (6.6) 131 (18.4)

Friend/peer 950 (51.6) 370 (51.7) Labourer 213 (11.8) 3 (0.4)

Outreach worker 280 (15.2) 148 (20.7) Sex worker 1 (0.1) 94 (13.2)

Peer educator 287 (15.6) 95 (13.3) Selling drugs 2 (0.1) 3 (0.4)

Family 207 (11.2) 71 (9.9) Other 256 (14.2) 87 (12.2)

Nurse 6 (0.3) 1 (0.1)

Other 111 (6.0) 30 (4.2)

* Percentages were calculated excluding missing cases

Table 2 OST treatment retention and outcomes over one year for a cohort of clients enrolled in May 2006 (n = 713)

Retained on OST Ceased – completed the program Ceased – relapsed Ceased – reason unknown

n (%) n (%) n (%) n (%)

3 months 519 (72.8) 18 (2.5) 138 (19.4) 38 (5.3)

6 months 451 (63.3) 42 (5.9) 166 (23.3) 54 (7.6)

9 months 405 (56.8) 60 (8.4) 186 (26.1) 62 (8.7)

12 months 362 (50.8) 91 (12.8) 196 (27.5) 64 (9.0)


of 7

complex setting where injecting drug use and a conse-quent HIV epidemic present substantial public healthchallenges. Previous studies have highlighted the positiveimpact of OST on reducing HIV risk behaviours, andimproving the quality of life and health of IDUs, and thefindings of this study also support the efficacy of OST asan intervention for people with opioid drug dependence.

HIV risk behaviours and socio-economic outcomesA range of behavioural, social and economic benefitswere evident as early as three months into treatment,and many of these benefits extended beyond the OSTclients to their families and communities. An importantoutcome of the OST program was a substantial reduc-tion in reported HIV risk behaviours, primarily unsafeinjecting and unsafe sex. Participation in the programarguably reduced the risk of HIV transmission not onlyfor those attending the program but also for their sexualpartners (often wives) and children. Other studies havealso found that OST is associated with rapid reductionsin HIV risk behaviours [3,5,6,17].Less family conflict has multiple positive flow-on

effects, especially for children. Fewer episodes of deten-tion or imprisonment reduces exposure to HIV risks,and suggests that less crime is being committed, animportant social outcome. One potential benefit of OSTprograms not evident in this study is an increase inemployment for the clients. This may be due to the factthat meaningful employment opportunities for relativelywell-educated young people in north-east India areextremely limited. Additionally, many (male) drug usersin north-east India are cared for by their natal families,

so are not forced to do menial work in order to obtainthe basic necessities of life.

Retention in treatmentRetention in OST treatment in Manipur and Nagaland(63% after six months) is comparable with retentionoutcomes reported by a WHO collaborative study thatincluded sites from low, middle and high-income coun-tries (approximately 70% after six months overall - only55% in Australia) [3]. Retention in treatment is clearlyimportant for the success of OST programs. As thefindings from this research and other studies indicate,the longer people are retained in an OST program, thegreater the likelihood that they will complete the pro-gram rather than relapse [17].Other studies have reported buprenorphine dose as an

important determinant of retention in the treatmentprogram (higher doses being associated with betterretention) [1]. While information about the dose ofbuprenorphine at the point of cessation was recordedfor some of the clients in this study, the extent of miss-ing data for this variable precluded meaningful analysis.

Implications for policies and programsAlmost half of the clients who ceased OST did so hav-ing completed the program without returning to theirprevious pattern of opiate drug use at the time of dis-charge, whilst the other half ceased OST due to relapse.Factors that significantly increased the likelihood ofceasing treatment due to relapse were higher spendingon drugs at intake, frequently missing buprenorphinedoses, and reporting heroin as their most problematic

Table 3 Changes in HIV risk behaviours when intake is compared with three months after enrolment (disaggregatedby status of client at the end of the data collection period)

Intake 3 months p-value*

Had shared a needle during past month (%)

Completed the program (n = 297) 23.5 0.7 <0.001

Relapsed (n = 155) 25.8 1.3 <0.001

Still on OST (n = 847) 27.6 2.1 <0.001

All clients (n = 1299) 26.5 1.8 <0.001

Had an unsafe sexual encounter during past month (%)

Completed the program (n = 260) 14.6 9.6 0.11

Relapsed (n = 138) 15.4 4.3 0.01

Still on OST (n = 818) 15.5 7.6 <0.001

All clients (n = 1216) 15.3 7.6 <0.001

Had been jailed/detained during past month (%)


Relapsed (n = 155) 12.9 0.0 <0.001

Still on OST (n = 841) 11.7 1.1 <0.001

All clients (n = 1293) 11.6 0.8 <0.001

* McNemar’s Test


of 7

drug. Longer duration in treatment and regular familyinvolvement significantly decreased the likelihood ofrelapse.Strategies to strengthen retention rates need to be

identified and implemented in order to achieve success-ful outcomes for a larger proportion of clients. Facilitat-ing family involvement in OST treatment could help toachieve better outcomes for clients, as might active fol-low-up and additional support for clients who are regu-larly missing buprenorphine doses and for those whoidentified heroin (rather than SP) as their most proble-matic drug. Additionally, given that a large proportionof clients who relapse leave the program in the first fewweeks, more intensive support for clients over the firstmonths of treatment may be beneficial.

Clients classified as “completing the program” werenot necessarily totally abstinent, but were no longerrequiring buprenorphine and had not returned to theirformer pattern of drug use at the time of discharge.Substantial reductions in drug use and HIV risk beha-viours should be the goal of OST, rather than absti-nence [17]. A systematic review of published researchfrom 1966 to 2003 reported that post-treatment absti-nence rates varied between 22% and 86%; overall 33% offormer OST (methadone) patients were abstinent fromat least opioids for an average of more than two yearsafter completing detoxification [18]. Another review ofOST research conducted in Germany (methadone)reported that only 10% of clients became totally absti-nent, and identified the concern that attempts to intro-duce time-limited (abstinence-oriented) treatment wouldresult in relapse and physical and psychological instabil-ity [19]. It is probable that the situation is similar innorth-east India i.e. only a small proportion of OST cli-ents are likely to achieve long-term total abstinence.Many will need to remain in the program for years, andsome will require lifelong treatment.

LimitationsAn important limitation of this analysis is that the datawere based primarily on self-report measures. Socialacceptability bias may have influenced the IDUs tounderstate the extent to which they were engaging inHIV risk behaviours, particularly given that the datawere being collected by program staff, though the com-parability of the results with controlled trials suggeststhat any effect of this kind may have been minimal. TheOST program in north-east India was better resourcedwhen funded by DFID than is currently the case (OSTis now funded by government), so we cannot assumethat current outcomes are the same as those reported inthis study.The length of follow-up is too short to draw any firm

conclusions about longer-term outcomes for these OSTclients. A prospective longitudinal cohort study to sys-tematically follow OST clients for 1-2 years would pro-vide valuable information about outcomes for OSTclients, the impact of various dosing schedules, socialand economic benefits, program costs, and the extent towhich clients are cycling in and out of the program.Additionally, it would be useful to follow clients whocease treatment (for whatever reason) to compare thebenefits of staying in treatment over those of leaving. Aqualitative investigation to follow-up clients who relapse,in order to better understand their reasons for relapse,would contribute to more effective programming.The findings from this study indicate that this OST

program in Manipur and Nagaland, which was imple-mented by NGOs in a severely constrained context

Table 4 Changes in quality of life indicators when intakeis compared with three months after enrolment(disaggregated by status of client at the end of the datacollection period)

Intake 3months

p-value*

Clients reporting a good quality of life(%)


Relapsed (n = 155) 17.4 54.8 <0.001

Still on OST (n = 849) 13.0 63.5 <0.001

All clients (n = 1301) 13.8 63.0 <0.001

Employed (%)

Completed the program (n = 297) 52.5 51.2 0.69

Relapsed (n = 155) 53.5 47.1 0.11

Still on OST (n = 844) 53.8 52.6 0.52

All clients (n = 1296) 53.5 51.6 0.17

Days in family conflict during pastmonth (mean)

Completed the program (n = 293) 4.6 0.5 <0.001+

Relapsed (n = 152) 4.1 0.8 <0.001+

Still on OST (n = 833) 4.6 0.7 <0.001+

All clients (n = 1278) 4.5 0.6 <0.001+

Social events attended during pastmonth (mean)


Relapsed (n = 155) 1.3 1.8 0.03+

Still on OST (n = 841) 1.3 2.0 <0.001+

All clients (n = 1293) 1.3 1.9 <0.001+

Day absent from work during pastmonth (mean)


Relapsed (n = 62) 1.9 1.0 0.22+

Still on OST (n = 350) 2.8 1.2 <0.001+

All clients (n = 538) 2.6 1.0 <0.001+

* McNemar’s Test performed unless otherwise stated

+ Paired t-test


of 7

managed to achieve outcomes that are internationallycomparable. It has arguably made an important contri-bution to HIV prevention in the region, as well asimproving the quality of life for a large group of peoplewith opioid dependence, their families and communities.

AcknowledgementsMs Rachel Kabi, Mr Surmick Waribam, Project ORCHID staff, Project ORCHIDNGOs, Dr Suresh Kumar, Dr Richard Dinatale, and Associate Professor PeterDeutschmann who were all connected to the design and implementation ofthe OST program. Ms Kerryn O’Rourke previously undertook some analysis ofthe OST data assisted by Dr Richard DiNatale and Dr Priscilla Robinson, andthis report builds on that work. Support for the analysis was provided by theBill & Melinda Gates Foundation (through their funding of the North-eastIndia Knowledge Network grant and Project ORCHID), but the viewsexpressed herein are those of the authors and do not necessarily reflect theofficial policy or position of the Foundation.

Author details1Nossal Institute for Global Health, University of Melbourne, Victoria,Australia. 2National AIDS Control Organisation - Guwahati Office, Ministry ofHealth & Family Welfare, Government of India. 3Project ORCHID, EmmanuelHospital Association, Guwahati, India.

Authors’ contributionsAll authors contributed to interpretation of the findings and development ofthe manuscript. GA and MK undertook the statistical analysis and wrote thefirst draft of the manuscript. BL and CS provided expertise on the context oflocal program delivery. NC provided expertise on opioid substitution therapyand harm reduction. All authors read and approved the final manuscript.

Authors informationCS was the OST Project Coordinator for Manipur and Nagaland during thecollection of data.

Competing interestsThe authors declare that they have no competing interests.

Received: 11 July 2010 Accepted: 1 December 2010Published: 1 December 2010

References1. Mattick RP, Kimber J, Breen C, Davoli M: Buprenorphine maintenance

versus placebo or methadone maintenance for opioid dependence.Cochrane Database Syst Rev 2004, CD002207.

2. Johnson RE, McCagh JC: Buprenorphine and naloxone for heroindependence. Curr Psychiatry Rep 2000, 2:519-526.

3. Lawrinson P, Ali R, Buavirat A, Chiamwongpaet S, Dvoryak S, Habrat B, Jie S,Mardiati R, Mokri A, Moskalewicz J, et al: Key findings from the WHOcollaborative study on substitution therapy for opioid dependence andHIV/AIDS. Addiction 2008, 103:1484-1492.

4. Connock M, Juarez-Garcia A, Jowett S, Frew E, Liu Z, Taylor RJ, Fry-Smith A,Day E, Lintzeris N, Roberts T, et al: Methadone and buprenorphine for the

management of opioid dependence: a systematic review and economicevaluation. Health Technol Assess 2007, 11:1-171, iii-iv.

5. Hubbard RL, Craddock SG, Anderson J: Overview of 5-year followupoutcomes in the drug abuse treatment outcome studies (DATOS). JSubst Abuse Treat 2003, 25:125-134.

6. Gowing L, Farrell M, Bornemann R, Sullivan L, Ali R: Substitution treatmentof injecting opioid users for prevention of HIV infection. CochraneDatabase Syst Rev 2008, CD004145.

7. WHO, UNODC, UNAIDS: Technical guide for countries to set targets foruniversal acess to HIV prevention, treatment and care for injecting drugusers. 2009.

8. Thirthalli J, Chand PK: The implications of medication development in thetreatment of substance use disorders in developing countries. Curr OpinPsychiatry 2009, 22:274-280.

9. Sharma M, Oppenheimer E, Saidel T, Loo V, Garg R: A situation update onHIV epidemics among people who inject drugs and national responsesin South-East Asia Region. Aids 2009, 23:1405-1413.

10. Mathers BM, Degenhardt L, Ali H, Wiessing L, Hickman M, Mattick RP,Myers B, Ambekar A, Strathdee SA: HIV prevention, treatment, and careservices for people who inject drugs: a systematic review of global,regional, and national coverage. Lancet 2010, 375:1014-1028.

11. Chandrasekaran P, Dallabetta G, Loo V, Rao S, Gayle H, Alexander A:Containing HIV/AIDS in India: the unfinished agenda. Lancet Infect Dis2006, 6:508-521.

12. Eicher AD, Crofts N, Benjamin S, Deutschmann P, Rodger AJ: A certain fate:spread of HIV among young injecting drug users in Manipur, north-eastIndia. AIDS Care 2000, 12:497-504.

13. National AIDS Control Organisation: Annual sentinel surveillance countryreport 2006. National AIDS Control Organisation. 2007.

14. UNODC: Drug use in the northeastern states of India. UNODC. 2006.15. Kumar MS, Natale RD, Langkham B, Sharma C, Kabi R, Mortimore G: Opioid

substitution treatment with sublingual buprenorphine in Manipur andNagaland in Northeast India: what has been established needs to becontinued and expanded. Harm Reduct J 2009, 6:4.

16. Zachariah R, Harries AD, Ishikawa N, Rieder HL, Bissell K, Laserson K,Massaquoi M, Van Herp M, Reid T: Operational research in low-incomecountries: what, why, and how? Lancet Infect Dis 2009, 9:711-717.

17. Ward J, Mattick R, Hall W: Methadone Maintenance Treatment and OtherOpioid Replacements Therapies. Amsterdam: Harwood AcademicPublishers; 1998.

18. Kornor H, Waal H: From opioid maintenance to abstinence: a literaturereview. Drug Alcohol Rev 2005, 24:267-274.

19. Michels I, Stover H, Gerlach R: Substitution treatment for opioid addicts inGermany. Harm Reduct J 2007, 4:5.

doi:10.1186/1477-7517-7-29Cite this article as: Armstrong et al.: Opioid substitution therapy inmanipur and nagaland, north-east india: operational research in action.Harm Reduction Journal 2010 7:29.

Table 5 Binary logistic regression model to predict the likelihood of relapse from OST treatment (n = 895)

Variable Unadjusted Odds Ratio (95% C.I.) p-value Adjusted Odds Ratio (95% C.I.) p-value

Male 1.22 (0.74, 2.02) 0.44 0.82 (0.34, 2.01) 0.67

Age (years) 1.01 (0.99, 1.04) 0.20 1.01 (0.97, 1.05) 0.63

Duration in treatment (months) 0.74 (0.71, 0.77) <0.001 0.76 (0.72, 0.80) <0.001

Heroin as the most problematic drug (ref: SP) 1.31 (0.96, 1.78) 0.09 1.95 (1.16, 3.28) 0.01

Money spent daily on drugs at intake (Rs 100 units) 1.18 (1.05, 1.33) 0.01 1.20 (1.00, 1.44) 0.05

Frequently missed more than two doses a week 14.67 (9.21, 23.35) <0.001 8.82 (4.99, 15.63) <0.001

Regular family involvement in treatment 0.11 (0.08, 0.15) <0.001 0.20 (0.13, 0.30) <0.001


of 7

http://www.ncbi.nlm.nih.gov/pubmed/15266465?dopt=Abstract




































opioid substitution therapy in manipur and nagaland, north-east india: operational research in...

Documents